New version page

RELIABILITY AND VALIDITY

Upgrade to remove ads

This preview shows page 1 out of 3 pages.

Save
View Full Document
Premium Document
Do you want full access? Go Premium and unlock all 3 pages.
Access to all documents
Download any document
Ad free experience

Upgrade to remove ads
Unformatted text preview:

RELIABILITY AND VALIDITY Reliability refers to the extent to which a measure is free from random error, and is consistent when measuring the same phenomenon over time. The following are some different methods used to assess reliability: • Test-Retest Reliability: the extent to which scores on the same measure, administered at two different times, correlate with each other. • Interrater Reliability: the extent to which two or more judges or raters agree on a given measure or assessment. • Internal Consistency Reliability: the extent to which the scores on the items of a scale correlate with each other. Usually assessed using coefficient alpha. • Equivalent-Forms Reliability: the extent to which scores on similar, but not identical, measures, administered at two different times, correlate with each other. Validity generally refers to the extent to which a measure accurately assesses the specific phenomenon that the researcher is attempting to measure. • Construct Validity: the degree to which a measured variable actually measures the conceptual variable that it is designed to measure. o Content Validity: the extent to which the measured variable appears to have adequately covered the full domain of the conceptual variable. (E.g., an intelligence test that contained only mental rotation tasks would be sorely lacking in content validity.) o Convergent Validity: the extent to which a measured variable is found to be related to other measured variables designed to measure the same conceptual variable. o Discriminant Validity: the extent to which a measured variable is found to be unrelated to other measured variables designed to measure other conceptual variables. • Criterion Validity: the extent to which a measure correlates with a behavioral measured variable (e.g., a self-report measure of shyness would be lacking in criterion validity if people who scored high on it were also observed to be outgoing). o Predictive Validity: the extent to which a measure correlated with (predicts) a future behavior o Concurrent Validity: the extent to which a measure correlates with a behavior measured at the same time• Internal Validity: the extent to which experimental results can be attributed to the manipulation of the independent variable. (Note that there are a number of threats to internal validity; see below). • External Validity: the extent to which results can be generalized to other populations and settings. THREATS TO VALIDITY History. Any thing that has occurred to the subjects between Time1 and Time 2 in addition to the independent variable may affect the dependent variable. Example: in addition to the psychotherapy treatment she received (the independent variable), a manic-depressive subject may view a television program on bipolar illness, which influences her behavior. Maturation. Subjects may change from Time 1 to Time 2 because of autonomous growth or development. Example: a subject who is rated as hyperactive at Time 1 (age 3) may grow out of his overactivity by Time 2 (age 4). The problems of history and maturation increase the longer the time period between Time 1 and Time 2. Testing (Practice Effects). Taking the pre-test itself may affect the dependent variable. Example: taking a practice test for the SAT before an educational intervention (the independent variable) not only assesses the subjects’ pre-treatment skills, but also teaches them something about test-taking and would improve their post-treatment performance even without the treatment. Instrument Change. The measuring instruments may change between Time 1 and Time 2. Example: during a longitudinal study, it is likely that the DSM diagnostic criteria will have been revised, resulting in changes in diagnosis that are not due to the independent variable. Statistical Regression to the Mean. Regression to the mean is expected of extreme scores from Time 1 to Time 2. This has nothing to do with the effect of the independent variable. Example: people who do extremely poorly on an IQ test at the first testing are likely to do somewhat better on the second test (and vice versa). Selection. Non-random selection may influence the independent variable. Example: if subjects are self-selected for an SAT course, those who take the course are likely to vary from the no-treatment group on some factor (like achievement motivation) in addition to the independent variable (SAT course). Experimental Mortality. (refers to those who drop out of the study) Non-random loss of subjects influences the independent variable: schizophrenic subjects who drop out of a medication treatment study may have been responding less to the treatment (causing them to quit); the resulting post-treatment group will not include those subjects who responded the least and consequently results may be somewhat inflated.References Davison, G. C., Neale, J. M., & Kring, A. M. (2004). Abnormal psychology (9th ed.). Hoboken, NJ: John Wiley & Sons, Inc. Davison, G. C., Neale, J. M., & Kring, A. M. (2004). Instructor’s resource manual for abnormal psychology (9th ed). Hoboken, NJ: John Wiley & Sons, Inc. Reber, A. S. (1995). Dictionary of psychology (2nd ed.). London: Penguin Books. Stangor, C. (2007). Research methods for the behavioral sciences. Boston: Houghton


Download RELIABILITY AND VALIDITY
Our administrator received your request to download this document. We will send you the file to your email shortly.
Loading Unlocking...
Login

Join to view RELIABILITY AND VALIDITY and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view RELIABILITY AND VALIDITY 2 2 and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?